Still more code to make sure we send the right number and kind of RELAY END cells

svn:r3723

doc/TODO

@@ -72,8 +72,8 @@ R . HTTPS proxy for OR CONNECT stuff. (For outgoing SSL connections to
   o Changes for forward compatibility
     o If a version is later than the last in its series, but a version
       in the next series is recommended, that doesn't mean it's bad.
-N . Do end reasons better
-    . Start using RESOURCELIMIT more.
+  o Do end reasons better
+    o Start using RESOURCELIMIT more.
     o Try to use MISC a lot less.
       o bug: if the exit node fails to create a socket (e.g. because it
         has too many open), we will get a generic stream end response.
@@ -82,11 +82,11 @@ N . Do end reasons better
         relay end feature.
     o Realize that unrecognized end reasons are probably features rather than
       bugs. (backport to 009x)
-    - Check for anyplace where we can close an edge connection without
+    o Push the work of sending the end cell deeper into package_raw_inbuf.
+      (Turns out, if package_raw_inbuf fails, it *can't* send an end cell.)
+    o Check for any place where we can close an edge connection without
       sending an end; see if we should send an end.
-    - Get some kind of sane return code from package_raw_inbuf, or maybe
-      push the work of sending the end cell deeper into package_raw_inbuf.
-N . Feed end reason back into SOCK5 as reasonable.
+N   . Feed end reason back into SOCK5 as reasonable.
 R o cache .foo.exit names better, or differently, or not.
 N - make !advertised_server_mode() ORs fetch dirs less often.
 N - Clean up NT service code even more.  Document it. Enable it by default.